Alibaba Unveils Qwen LLMs in China's Latest Open-Source AI Innovation

Alibaba released the next generation of its open-source large language models, Qwen3, on Tuesday, and experts are hailing it as yet another milestone in China's thriving open-source artificial intelligence space.

In a blog post, the Chinese tech giant stated that Qwen3 promises improvements in reasoning, instruction following, tool usage, and multilingual tasks, competing with other top-tier models such as DeepSeek's R1 in several industry benchmarks.

The LLM series includes eight variations that cover a wide range of architectures and sizes, giving developers more options when building AI applications for edge devices such as mobile phones.

Qwen3 also marks Alibaba's foray into so-called "hybrid reasoning models," which it claims combine traditional LLM capabilities with "advanced, dynamic reasoning."

According to Alibaba, such models can seamlessly switch between a "thinking mode" for complex tasks like coding and a "non-thinking mode" for faster, all-purpose responses. 

"Notably, the Qwen3-235B-A22B MoE model significantly lowers deployment costs compared to other state-of-the-art models, reinforcing Alibaba's commitment to accessible, high-performance AI," the company explained. 

Individual users can already access the new models for free on platforms such as Hugging Face, GitHub, and Alibaba Cloud's web interface. Qwen3 is also used to power Alibaba's artificial intelligence assistant, Quark.